Bipolar Depression at Nine Weeks' Gestation: Weighing Relapse Against Teratogenicity

Antidepressant monotherapy is the wrong answer in bipolar I disorder.

• Antidepressant monotherapy carries a substantial risk of treatment-emergent mania/mixed states and rapid cycling, and in pregnancy the consequence is not just symptom distress - it is hospitalization, impaired self-care, suicidality, and obstetric risk. A prior manic episode with psychosis is a strong marker of severe illness.
• If an antidepressant is used at all, it must be adjunctive to a mood stabilizer and preferably not a tricyclic (higher switch risk).

Her relapse risk is high: untreated bipolar illness in pregnancy is associated with high rates of recurrence, especially after stopping lithium and with a history of severe mania.

Lithium recommendation:

• Re-initiate lithium after shared decision-making, with the absolute risk framed properly: the classic estimate for Ebstein anomaly is roughly 1 in 1,000 exposed pregnancies versus a background of about 1 in 20,000 - an increase in absolute terms of a fraction of a percent, and recent cohort data suggest a smaller effect than originally reported. The absolute relapse risk from withholding treatment is far larger.
• Give folic acid 5 mg daily, arrange a fetal echocardiogram at 18–20 weeks and a level II anatomy scan, and coordinate with obstetrics.
• Target a trough level of 0.6–0.8 mmol/L, drawn 12 hours after the dose; check weekly after initiation, then every 2–4 weeks, and more often near delivery. Lithium clearance rises in pregnancy (GFR increase) and falls sharply postpartum - dose requirements typically increase then drop abruptly after delivery.
• Monitor creatinine, TSH, and calcium each trimester; emphasize hydration and NSAID avoidance.
• Have a postpartum plan in writing (level check within 24–48 hours of delivery, dose reduction, sleep protection, psychiatric follow-up within a week).
• Document capacity, the discussion, and her preferences; she is not acutely psychotic and is participating in decisions.

The level of 2.1 mcg/mL is a pre-pregnancy value and cannot be assumed to still apply.

• Lamotrigine clearance increases substantially in pregnancy - commonly 2-fold or more by the third trimester, driven by estrogen-induced glucuronidation (UGT1A4). Levels fall progressively, and the greatest risk is in the second and third trimesters.
• Obtain a current level now and then monthly, with dose increases as needed to maintain the pre-pregnancy therapeutic concentration. Do not simply accept a low level as "expected."
• Titrate upward slowly - rapid re-escalation risks rash including Stevens-Johnson syndrome. Note that the pregnancy-era dose after upward titration is usually tolerated well, but the same dose is excessive postpartum.
• Postpartum plan: reduce back toward the pre-pregnancy dose over roughly 2–3 weeks after delivery with level guidance; the falling estrogen level otherwise produces toxicity (diplopia, ataxia, sedation).

Why lamotrigine alone is not enough here:

• Lamotrigine has weak antimanic and prophylaxis-of-mania activity - it is a depression-predominant agent. This patient's defining event was psychotic mania.
• She is already depressed on lamotrigine plus low-dose quetiapine, so the current regimen is failing.

Practical sequence: hold the antidepressant question on the shelf until she is on an adequate mood stabilizer; treat the depression with lithium ± optimized quetiapine; reassess in 2–3 weeks with a repeat MADRS and a suicide risk assessment at every contact.

Quetiapine 50 mg at bedtime is a sleep dose, not an antidepressant dose.

• For bipolar depression, quetiapine is effective at 300 mg/day (immediate- or extended-release), with 150–300 mg in the 150–300 range showing benefit. 50 mg will not treat a MADRS of 27.
• If increased, counsel on orthostasis and sedation (fall risk, driving), and monitor weight, glucose/A1c, and lipids each trimester - pregnancy itself is a metabolic stressor and she may already be gaining weight.
• Quetiapine is a reasonable pregnancy choice (no consistent major malformation signal; extensive data), and it can serve as an adjunct if lithium alone is insufficient.

Postpartum - plan it during the second trimester, not after delivery:

1. Bipolar illness carries the highest relapse risk in the first weeks postpartum, and sleep deprivation is a potent trigger. Arrange sleep protection (partner night feeds if formula/bottle supplementation is acceptable to her), frequent psychiatric contact, and a written crisis plan.
2. Lithium level within 24–48 hours of delivery and dose reduction per renal clearance; watch for toxicity if she is dehydrated or post-cesarean with fluid shifts.
3. Lactation: lithium is measurable in breast milk and requires infant serum levels, thyroid and renal checks if breastfeeding - many clinicians advise against it, but it can be done with informed consent and monitoring. Quetiapine and olanzapine have low relative infant dose and are preferred if an antipsychotic is needed while breastfeeding; monitor the infant for sedation and weight gain.
4. Lamotrigine is generally compatible with breastfeeding, but the infant receives a small dose - watch for rash.
5. Give her written guidance on postpartum warning signs (decreased sleep need, racing thoughts, irritability) and a same-day contact route.